We will use molecular approaches in three experimental contexts to define the relationship of hepatitis B viruses (HBV)-like agents to cancer. 1. The genomes of two HBV-like agents of rodents, the woodchuck hepatitis virus (WHV) and the ground squirrel hapatitis virus (GSHV), will be recombined to seek determinants of the different host range and pathological potential of these two viruses in studies that exploit our demonstrated ability to establish infections with cloned viral DNA. 2. To investigate the role of viral genes in hepatoma development, we will determine the organization and expression of HBV DNA in primary human hepatomas. These tumors will also be examined for the presence of genetic lesions manifested by (a) HBV insertion mutations that repeatedly affect the same host gene; (b) aberrant structure or expression of known oncogenes; or (c) biologically-active oncogenes capable of transforming NIH/3T3 or cultured hepatic cells. 3. The ability of the DNA of HBV-type viruses to serve as activators (enhancers) of linked heterologous genes will be studied in cell culture systems.